• Social (im)mobility

    TIE readers might appreciate a piece I posted on Friday in the Washington Post’s Wonkblog section. I raise a few quibbles regarding Greg Mankiw’s recent philosophical analysis of inequality. Along the way, I cite an interesting empirical analysis of intergenerational wealth mobility by Duke sociologist Lisa Keister in her 2005 book, Getting rich: America’s new rich and how they got that way.

    My Wonkblog  column doesn’t provide useful specific detail, but Keister’s table 2.10 provides food for thought. She examines wealth among respondents in the National Longitudinal Survey of Youth (NLSY79). She examines mobility across socioeconomic quintiles among respondents who were living with their parents in 1979, and were (almost all) living on their own in the year 2000:

    Adult quintile year in the year 2000

    Childhood quintiles in 1979




































    As I read the table, there isn’t a whole lot of intergenerational mobility here. A child living in a household at the top quintile has a 55.1% chance of remaining in the top quintile, and only an 11% chance of ending up in the bottom 40%. The comparable numbers for someone in the middle quintile are 13% and 31%, respectively. And of course movement out of the bottom quintile is pretty much the mirror-image of movement out of the top.

    It’s good to be born right.

    • What would the chart look like in a perfect meritocracy with equal opportunity? It wouldn’t show no correlation between SES at birth and at age 40, since intelligence and focus can be inherited and successful, intelligent parents tend to teach their children good work habits and offer intellectual stimulation.

    • Paul Krugman on the subject: http://krugman.blogs.nytimes.com/2013/06/22/greg-mankiw-and-the-gatsby-curve/ He links to some good analyses.

    • Harold, I agree that income and wealth inequality is a critical economic, social, political, and moral problem in the United States today, and I fully support federal and state government policies to ameliorate this predicament. However, I do not find Keister’s table 2.10 to be as discouraging as you seem to do.

      I notice that each childhood quintile has a 20-25% chance of moving up to the next higher quintile (except the top childhood quintile which cannot move up) and each has a 20-25% chance of moving down to the next lower quintile (except the bottom childhood quintile which cannot move down).

      The middle childhood quintile has a 34.2% chance of moving up to a higher quintile and a 30.71% chance of moving down to a lower quintile.

      The bottom childhood quintile has a 54.7% chance of moving up to a higher quintile and the top childhood quintile has a 44.8% chance of moving down to a lower quintile.

      All this seems to me to show a significant degree of social mobility, both up and down the socio-economic ladder. Also, the nature of this table makes it mathematically impossible for everybody to move up (or down) — that is, everybody cannot move into the top quintile because by definition only 20% of the population can be in the top quintile. Finally, I believe some (arguable degree of) social immobility is due to parentally (both genetically and culturally) transmitted talents and values. This makes it marginally more likely that a child will remain in the same quintile as his/her parents even in the face of ideal government policies.

      Nevertheless, does this table represent the ideal degree — or even a tolerable degree — of social mobility? No, that’s why I still support government health, nutritional, and educational support programs, among others. What other data is available to measure social mobility/immobility and elucidate its causes?

    • “A new report from the Organization for Economic Co-Operation and Development (OECD) finds that social mobility between generations is dramatically lower in the U.S. than in many other developed countries.

      So if you want your children to climb the socioeconomic ladder higher than you did, move to Canada.” http://www.huffingtonpost.com/2010/03/17/social-immobility-climbin_n_501788.html

      Referenced OECD Report:

    • “It’s good to be born right.” Okay, but what’s the point? Bigger and more centralized government? More redistribution of income? Mr. Pollack neglected to make any policy recommendations.

      By the way, Mr. Pollack seems to read the evidence in a way perhaps that supports his preconceived ideas. In Keister’s table, more than half of the lowest quintile moved up and almost half of the top quintile moved down. Seems like significant mobility to me.

    • Mankiw is very likable. I particularly enjoyed his commencement speech at this son’s high school graduation that he posted on his web site. His problem, as I see it, is one of framing, which is very much from a right and a left. I would contrast Mankiw’s framing with Robert Shiller’s. Shiller is an economics professor at Yale, best known in matters of finance. His lectures on finance provide a very different framing of the subject. To Shiller, finance is all about risk management. That’s intuitive, I suppose, but he goes much farther, by describing the perfect world as one in which all risks are shared. Sure, moral hazard is a big part of Shiller’s framing, but to Shiller, moral hazard is simply a problem to be solved by creative minds trying to improve the world by sharing risk. To Mankiw, moral hazard is a wall, the other side of which is dystopia. Shiller is the positivist, Mankiw is the negativist.

    • It is hard for me to imagine a cause for this other than a cultural or genetic transmission of diligence and ambition (and to to a much lesser extent genetic transmission of intelligence).

      And if a less diligent and ambitious culture exists among the people of the lower quintiles who are we to attempt to change that?

      For example maybe people in the bottom quintile are more likely to say the child does enough school work at school I let him enjoy his childhood when not in school.

      I am amazed when I hear that some child of a rich person earns good income from his work (outside a family business or earnings off investments). Yet many children of the rich work quite hard. Why they do not need the money.

      • The notion that the economic immobility of those born in the lower quintile *must* be due to some inherent moral failing of themselves, their parents, and/or the culture of their peers sounds like typical calvinist American classism–swap out a few social science colloquialisms and its little better than our grandfathers’ racism. I’m sure its very comforting for the privileged to attribute their success to their own greatness, and the failings of others to their own inferiority–its a shame we can’t all be as great as you think you are, oh diligent and ambitious one!

        Surely one’s outcomes are at least in part related to the opportunities and protections afforded to them by the institutions that comprise our society? Things like access to safe living environments, nutrition, education, and (perhaps most importantly) equal protection under the law? Could it be possible that the game is increasingly rigged to comfort the comfortable and afflict the afflicted? There’s been oceans of ink spilled by economists and other social scientists, as well as pundits and other media types in this and other countries, which strongly suggests that inequality is increasing rapidly, and that its not because of an epidemic of laziness amongst “those” people.

        SB in StL

    • “it’s good to be born right”

      more appropriately: it’s good to be born American. If this analysis was done for a global cohort, 1979 to 2000, I assume that almost all Americans would be in Q5 in both years.

      and, extended across all of history, all present Americans would be in the top 99% of humans that have EVER LIVED.

      be grateful!

      [posted from Nairobi]

    • Canada has more relative income mobility than the USA because it has more income equality. That is, the gap between, say, the 20th percentile of income and the 80th percentile is much smaller in Canada than the USA. That means it takes much smaller changes in income to “change quintiles” in Canada than in the USA.

      What this really means is that the supposedly higher income mobility in Canada is merely a statistical artifact of Canada’s higher income equality. Most of the “data” used to show that Canada, or other countries, have higher income mobility than the USA are really just another statistic telling us that they have less income inequality but is meaningless when looking at income mobility.

      It may well be true that Canada has much higher income mobility than the USA, but the data here doesn’t show that. I would argue that looking at absolute mobility would be a better (albeit still flawed) way of measuring. If anyone knows of data that allows us to compare absolute income growth across the income spectrum, that would be very helpful.